whole plant acclimation responses by finger millet to low ...€¦ · manish n. raizada* department...

ORIGINAL RESEARCHpublished: 19 August 2015

doi: 10.3389/fpls.2015.00652

Edited by:Raul Antonio Sperotto,

Centro Universitário Univates, Brazil

Reviewed by:Anil Kumar,

G. B. Pant University of Agricultureand Technology, India

David D. Baltensperger,Texas A&M University, USA

*Correspondence:Manish N. Raizada,

Department of Plant Agriculture,University of Guelph, 50 Stone Road

East, Guelph, ON N1G 2W1, [email protected]

Specialty section:This article was submitted to

Plant Nutrition,a section of the journal

Frontiers in Plant Science

Received: 27 May 2015Accepted: 06 August 2015Published: 19 August 2015

Citation:Goron TL, Bhosekar VK, Shearer CR,

Watts S and Raizada MN (2015)Whole plant acclimation responses by

finger millet to low nitrogen stress.Front. Plant Sci. 6:652.

doi: 10.3389/fpls.2015.00652

Whole plant acclimation responsesby finger millet to low nitrogen stressTravis L. Goron, Vijay K. Bhosekar, Charles R. Shearer, Sophia Watts andManish N. Raizada*

Department of Plant Agriculture, University of Guelph, Guelph, ON, Canada

The small grain cereal, finger millet (FM, Eleusine coracana L. Gaertn), is valued bysubsistence farmers in India and East Africa as a low-input crop. It is reported by farmersto require no added nitrogen (N), or only residual N, to produce grain. Exact mechanismsunderlying the acclimation responses of FM to low N are largely unknown, both aboveand below ground. In particular, the responses of FM roots and root hairs to N or anyother nutrient have not previously been reported. Given its low N requirement, FM alsoprovides a rare opportunity to study long-term responses to N starvation in a cerealspecies. The objective of this study was to survey the shoot and root morphometricresponses of FM, including root hairs, to low N stress. Plants were grown in pails in asemi-hydroponic system on clay containing extremely low background N, supplementedwith N or no N. To our surprise, plants grown without deliberately added N grew tomaturity, looked relatively normal and produced healthy seed heads. Plants respondedto the low N treatment by decreasing shoot, root, and seed head biomass. Thesedeclines under low N were associated with decreased shoot tiller number, crown rootnumber, total crown root length and total lateral root length, but with no consistentchanges in root hair traits. Changes in tiller and crown root number appeared tocoordinate the above and below ground acclimation responses to N. We discuss theremarkable ability of FM to grow to maturity without deliberately added N. The resultssuggest that FM should be further explored to understand this trait. Our observationsare consistent with indigenous knowledge from subsistence farmers in Africa and Asia,where it is reported that this crop can survive extreme environments.

Keywords: finger millet, nitrogen stress, grain, shoot, root, root hair, crown root, lateral root

Introduction

Finger millet (FM, Eleusine coracana L. Gaertn) is one of the small millet cereals (National ResearchCouncil, 1996; Goron and Raizada, 2015), originally native to the Ethiopian highlands (Dida et al.,2008). FM is largely consumed by marginalized inhabitants of semi-arid Asia and Africa, andsold to provide subsistence farmers with additional income (Dorosh et al., 2009; Gruère et al.,2009). FM is also highly valued by local farmers for its ability to grow in adverse agro-climaticconditions, where major cereal crops such as maize (Zea mays), wheat (Triticum spp.), and rice(Oryza sativa) fail, and has been noted to tolerate a wide variety of soils (Upadhyaya et al.,2006).

Abbreviations: DAT, days after transplant; FM, finger millet; NUE, nitrogen use efficiency.

Frontiers in Plant Science | www.frontiersin.org 1 August 2015 | Volume 6 | Article 652

http://www.frontiersin.org/Plant_Science/

http://www.frontiersin.org/Plant_Science/editorialboard

http://www.frontiersin.org/Plant_Science/editorialboard

http://dx.doi.org/10.3389/fpls.2015.00652

http://creativecommons.org/licenses/by/4.0/

http://dx.doi.org/10.3389/fpls.2015.00652

http://journal.frontiersin.org/article/10.3389/fpls.2015.00652/abstract

http://loop.frontiersin.org/people/200707




http://www.frontiersin.org/

http://www.frontiersin.org/Plant_Science/archive

Goron et al. Finger millet under nitrogen stress

Related to the latter point, FM is also valued for itsexceptionally high nitrogen use efficiency (NUE; Gupta et al.,2014), defined as grain yield per unit of available nitrogen (N;Moll et al., 1982). Compared to other grain crops such as maize,FM responds very well to low amounts of N (Roy et al., 2001).Many subsistence farmers in South Asia have reported to us thatFM can grow without any added N or with only residual N.A more formal study carried out by the All India CoordinatedSmall Millet Improvement Project (AICSMIP) indicated thatwhile FM responds well to urea N application at 90 kg/ha,the cost benefit ratio was highest between 0 and 30 kg N/ha1.This is a significantly lower N requirement than corn, forwhich optimum application rates can be greater than 200 kg/ha(Debruin and Butzen, 2014). Other work has confirmed theseobservations: a four year study in 2011 showed that FM grainyield stayed consistent between fertilizer application rates of20–40 kg N/ha (Pradhan et al., 2011). Limited research alsosuggests that FM genotypes vary in their NUE (Thilakarathna andRaizada, 2015), with some genotypes recognized as having higherresponsiveness to applied N (Puttaswamy and Krishnamurthy,1976; Bhoite and Nimbalkar, 1996; Dubley and Shrivas, 1999;Gupta et al., 2012, 2013, 2014). There is paucity of literatureconcerning development of new FM genotypes with high yieldpotential under low N application regimes (Thilakarathna andRaizada, 2015). However, in recent years more work has beenaccomplished in this area (Goron and Raizada, 2015). Breedingefforts can be facilitated by selection for traits associatedwith NUE.

Despite FM being recognized as a high NUE crop (Guptaet al., 2014), the underlying mechanisms are not well understood.Crop plants utilize a wide range of acclimation strategies tomitigate the limitations of low N availability, both morphologicaland biochemical, including altering root traits for betternutrient salvaging, decreasing chlorophyll production, changingN allocation within the plant, and altering the timing of flowering(Chapin et al., 1987; Ciampitti et al., 2013).

Particularly poorly characterized acclimation responses to lowN in plants include changes in root growth and architecture.The cereal root system consists of thick crown roots whichinitiate from the base of the stem (the crown region), fromwhich lateral roots extend and branch; all of these root types canfurther initiate root hairs, which are epidermal projections thatassist with nutrient uptake (Figure 1A). Maize exhibits diverseroot responses to low N, for example by increasing the totallength of the root system, decreasing the crown root number,and increasing the lateral root to crown root ratio (Eghballand Maranville, 1993; Wang et al., 2004; Chun et al., 2005; Liuet al., 2009; Gaudin et al., 2011a). To a lesser extent, low Nis associated with altered root hair length and density in cropgrasses (Gaudin et al., 2011a,b). Under extreme N deficiency(N starvation), Arabidopsis thaliana ecotypes show a range ofroot responses to N starvation (Ikram et al., 2012), though suchextreme experiments are more difficult to conduct with the majorcereals, because of their high N requirement for viability after theseedling stage.

1http://smallmillets.res.in/html/reports.html

To the best of our knowledge, there have been no reportsconcerning root architecture or root hair traits in FM except fora recent root hair methodology paper by our group (Goron et al.,2015). Furthermore, we could not find reports in the literatureconcerning other detailed morphological acclimation responsesof FM in response to low N. Given its low N requirement andsmall seed size as a source of starter N, FM may also provide arare opportunity to understand long-term acclimation responsesof a cereal to extreme N stress.

The objective of this study was to survey shoot androot morphometric acclimation responses of FM to very lowbackground N. To ensure minimal levels of N, plants were grownin pails containing an inert clay substrate called Turface R© in asemi-hydroponic system without added N (Tollenaar and Migus,1984; Figures 1B,C). This system permitted a more detailedanalysis of fine root traits including root hairs, as shown by ourgroup with maize (Gaudin et al., 2011a,b) and recently in FM(Goron et al., 2015), compared to excavation from soil. At thebeginning of the study, it was unclear whether FM plants wouldreach maturity in the absence of added N.

Materials and Methods

Plant Materials and Growth ConditionsA field experiment was carried out over two growing seasons(2012 and 2013). A commercial variety of Eleusine coracanaL. Gaertn (FM) seeds was obtained from India. Seeds weregerminated in open trays at room temperature, supplied withonly double distilled water (ddH2O) in a laboratory for 7 daysbefore transplantation into pails in a field near Guelph, Canada(43◦53′N, 80◦18′W, 325 m above sea level). Single FM plantswere grown in 22 L plastic pails (28 cm in diameter) containingTurface R© MVP (Profile Products LLC., Buffalo Grove, IL, USA).This is an inert, baked-clay, coarse growth medium which canbe used in automated field fertigation systems as previouslydescribed by Tollenaar and Migus (1984). Within the field, thedistance between the centers of pails in each row was 35 cm, andthe spacing between rows was 142 cm. Pails were arranged inblocks consisting of two N treatments (see below). There werefour blocks per row, with blocks distributed in the field acrossfive rows, for a total of 20 replicates per treatment per season.Each block was flanked by buffer plants which received varyingamounts of N. Plants were sampled randomly.

Millet plants were irrigated by an automated mechanism(Figure 1B), zero to three times per day, adjusted throughout thegrowing season as required. A concentrated, modified Hoagland’ssolution lacking N was stored in a 340 L reservoir at the field site,and was diluted to the appropriate concentration by the irrigationmechanism at a ratio of 1:100 during application. The pH of thediluted solution was adjusted to 6.5–6.7 by the addition of HCl.Two fertigation tubes calibrated to deliver a minimum of 10 mlof nutrient solution per minute were inserted into each pail (Pageet al., 2011). For the positive nitrogen treatment (+N), 1.1 g ofurea was dissolved into 1 L H20 (37 mM total N, to compensatefor leaching) and was provided to the plants at weekly intervals(13 times total) over the course of the experiment, with control


http://smallmillets.res.in/html/reports.html





FIGURE 1 | Guide to the FM root system and the semi-hydroponic fieldfertigation growth system used in this study. Diagram of a FM root network(A). Crown roots are the thick roots that initiate from the crown region at thebase of the shoot. Lateral roots initiate from the crown roots and cansubsequently branch. Root hairs are present on both crown roots (as shown)

and lateral roots. The fertigation growth system employed in this studyconsisted of 22 L plastic pails, 28 cm in diameter, filled with an inert baked claymedium (Turface R© ; B). Plants were allowed to grow to maturity as shown (C).Irrigation hoses delivered the nutrient solution except for nitrogen which wasadded manually as described.

plants receiving 1 L H2O with no nitrogen (−N) per dose.Nitrogen (or the water control) was supplied directly to eachexperimental unit by hand at the same time of day throughoutthe experiment.

Physiological, Seed Head, Shoot, and RootSystem MeasurementsChlorophyll levels of shoot tissue from a minimum of fiverandomly selected plants in each treatment were obtained witha Konica Minolta SPAD-502Plus chlorophyll meter (KonicaMinolta, Inc., Tokyo). Similar leaf positions were sampledbetween treatments. In 2012, one measurement was made at 106DAT; in 2013 two measurements were made at 50 and 80 DAT. Inboth years, plant flowering was recorded and tracked over time.A plant was considered to have flowered on the day at which theinflorescence was first visible. Tiller number was scored at 113DAT in 2012 and 119 DAT in 2013.

Seed head tissues were harvested for biomass analysis at 134DAT in 2012 and 135 DAT in 2013. For root and shoot biomassanalyses, at least 10 plants per treatment were collected at 142DAT in 2012 and 139 DAT in 2013. In 2013, three samplesof each tissue (10 g each of root, shoot, and seed heads) weredehydrated in a tissue dryer at 82◦C, ground to a fine powder inliquid nitrogen, and sent to the University of Guelph Agricultureand Food Laboratory for total N content quantification withthe Dumas combustion method (Fiedler et al., 1973) using aLECO FP428 nitrogen and protein determinator (LECO Corp.,St Joseph, MI, USA).

At the time of harvest in both years, five complete rootsystems from each treatment were frozen at −20◦C in 50%ethanol. Before analysis, roots were thawed inwater and floated in30 cm× 42 cm transparent plastic trays. Roots were scanned withan Epson Expression 10000XL large area scanner (Seiko EpsonCorporation, Suwa, Nagano, Japan), and the resulting images






were analyzed with WinRhizo software (Version PRO2009,Regent Instruments Inc., Québec, QC, Canada). The analysissoftware was set to measure total root length per diameter classallowing separate quantification of lateral roots (<0.45 mm) andcrown roots (>0.5 mm). Crown root number was scored bycounting visible roots in the crown region.

Root Hair Microscopy and MeasurementsFinger millet root hairs were measured using a protocol recentlydeveloped by our group (Goron et al., 2015). Briefly, root systemswere thawed by floating them in water. Four distinct crownroots of different lengths were selected from each root systemin order to obtain an accurate representation of the range ofcrown root ages within each root network. The four differentcrown root length classes were kept constant between plants. Fivesegments measuring 1 cm in length were dissected from each ofthe selected crown roots at equally spaced distances along theroot, and rinsed in double distilled water. Crown root segmentswere stained with 0.4% Trypan Blue solution (MP BiomedicalsLLC, Solon, OH, USA) for 10 min, and then rinsed five timeswith ddH2O. Root segments were subsequently immersed in 70%glycerol (Sigma-Aldrich, St. Louis, MO, USA), then examinedwith a Leica MZ8 stereomicroscope (Leica Microsystems GmbH,Wetzlar, Germany) under 5x magnification. Northern Eclipsesoftware (version 5.0, Empix Imaging Inc., Mississauga, ON,Canada) was used to capture four non-overlapping graphicimages of each 1 cm root segment with a Sony DXC-950P PowerHAD 3CCD color video camera (Tokyo, Japan). ImageJ software(Version 1.47,Wayne Rasband, NIH, USA) was used to manuallytrace root hairs to quantify root hair length and density by firstcalibrating the program to the image of a stagemicrometer (1 mmin length). For root hair length determination, all four imagesper crown root segment were used, with up to 10 root hairs ineach image traced. For statistical analyses, the mean root hairlength from all four images was calculated, representing a poolof up to 40 root hairs. In total, 14,420 root hairs were traced byhand for the experiment. For root hair density determination,for each crown root segment, a randomly selected 300 µm sub-segment was used to count the number of root hairs. For bothroot hair length and density measurements, there were a total offive replicate plants.

Determination of Potential Plant-AvailableNitrogen in the Clay Growth MediumA description of the physical-chemical properties of Turfaceclay is included (Supplementary Table S1). However, as nodata was available concerning the amount of N present in ourclay gravel growth medium, three replicates of 300 g groundTurface R© MVPwere sent to the University of Guelph Agricultureand Food Laboratory for total N content quantification, withthe Dumas combustion method (Etheridge et al., 1998). Thedry clay gravel was found to contain minimal levels of N(0.053%; Supplementary Table S2). Additionally, to determine Nbioavailability, 170 g of the growth medium was submerged for24 h in the same N-free nutrient solution provided to plants inthe field. The resulting slurry was filtered to remove the gravel,and three replicates of the filtrate were sent to the same lab

listed above for total N quantification by the Kjeldahl method(Keeney and Bremner, 1966). The filtrate was also found tocontain extremely low levels of N (1.42 mg/L total N, equivalentto 0.1 mM; Supplementary Table S2).

Statistical AnalysisAnalyses were performed using GraphPad Prism R© software(version 6.04; GraphPad Software, Inc., San Diego, CA, USA).ROUT was used to identify and remove outliers at Q= 1%. Therewas concern for the non-randomN treatments of the neighboringborder plants, though they were in separate pails. Nevertheless,to compensate for any position effects of neighboring plants onthe variance of each treatment, non-parametric Mann–Whitneytests were used to analyze datasets with non-normal values. Non-normality was identified with Shapiro–Wilk tests where replicatenumbers were sufficient and with Kolmogorov–Smirnov testswhere the sample size was small. Furthermore, unpaired t-testswith Welch’s correction for unequal standard deviations wereused to generate P-values.

Correlation coefficients were calculated by the two-tailedPearsonmethodwith a 95% confidence interval. Hypotheses wereevaluated at a type I error rate of 0.10 or 0.05 as indicated.

Results

Finger Millet Survival under N StarvationTo our surprise, plants that did not receive nitrogen (−N)deliberately at any point during the experiment (over a ∼140 dayperiod) reached maturity in both 2012 and 2013. The plantslooked normal, but smaller and less bushy, compared to plantstreated with N (+N), and even produced healthy seed heads(6.5 g per plant in 2012, and 13.7 g per plant in 2013; Figure 2D;Table 1).

Root, Shoot, and Seed Head BiomassIn both 2012 and 2013, end-season root and shoot biomassvalues were significantly lower when plants were treated with −Ncompared with +N (Tables 1 and 2; Figures 2 and 3).However, biomass values varied widely between years, with2012 plants being notably smaller than 2013 plants, a reflectionof differences in growing conditions between these years. Dueto these differences, as well as differences in methodology(i.e., recording dry shoot biomass in 2012 and fresh shootbiomass in 2013), biomass comparisons were restricted to withineach growing season, and no tests of significance across yearswere attempted. In 2012 and 2013, plants receiving the −Ntreatment showed 27 and 75% declines in seed head biomass,respectively, compared to plants that had received the +Ntreatment (Table 1).

Whole Plant ArchitectureAbove ground, in both years, the number of tillers per plantdecreased significantly when plants were treated with −Ncompared with +N, with a 30% decrease in 2012 and a60% decrease in 2013 (Table 1). Below ground, complete FMroot systems at maturity showed a high degree of fibrousness






FIGURE 2 | Finger millet shoot responses to low N at different growth stages. FM in 2012 at 7 weeks of growth (A) and at harvest (B). FM in 2013 at7 weeks of growth (C) and at harvest (D). All pails were 28 cm in diameter.

TABLE 1 | Measurements of above ground traits of FM in response to low nitrogen stress.

Year Treatment Duration Shoot biomass (g)∗ Seed head biomass (g)# Shoot tiller number

2012 +N 142 DAT 92.8 ± 2.23, n = 10 A 8.9 ± 0.43, n = 10 A 4.0 ± 0.26, n = 10 (113 DAT) A

−N 142 DAT 72.3 ± 0.80, n = 10 B 6.5 ± 0.27, n = 10 B 2.8 ± 0.20, n = 10 (113 DAT) B

2013 +N 139 DAT 869.9 ± 53.83, n = 20 a 54.9 ± 6.14, n = 20 a 50.5 ± 1.87, n = 13 (119 DAT) a

−N 139 DAT 168.0 ± 13.42, n = 20 b 13.7 ± 1.45, n = 20 b 19.9 ± 1.51, n = 13 (119 DAT) b

Letters that are different within a year indicate that the means are significantly different (P < 0.05).∗Dry biomass was measured in 2012, and fresh biomass in 2013.#Dry biomass was measured both years.

(Figure 3). In 2012, with the smaller plants, no measurementsof root architecture aside from biomass differed between Ntreatments (Table 2). In 2013, with the larger, more N-demandingplants, the number of crown roots on −N plants was 49% lessthan+N plants. Total crown root length, total lateral root length,and total root system lengths were all 37% smaller in −N plants(Table 2). There was no difference in terms of average crown rootlength or the ratio of total lateral to crown root length.

Correlation between Above Ground Biomassand Root CharacteristicsPearson correlations were performed between above groundbiomass and root characteristics using data collected in 2013(Figure 4; data from individual plants were not available in2012 to permit correlations). In 2013, a significant correlationwas observed between the numbers of crown roots and tillers(Figure 4A). The shoot fresh biomass also significantly andpositively correlated with the crown root number, total crownroot length and total lateral root length (Figures 4B–D,F), butnot with average crown root length (Figure 4E).

Additionally, Pearson correlations were performedbetween seed head dry biomass and the aforementionedroot characteristics for the 2013 season. There were positivecorrelations between seed head biomass and total root length,crown root number, total crown root length, and total lateral rootlength (Figures 4H–J,L), but not with average crown root length(Figure 4K), though some correlations were only significant atP = 0.10 (as indicated).

Combined, these data indicate that the declines in shoot andseed head biomass observed under severe low-N stress might be

matched by decreases in shoot tiller number as well as the totallengths of the two main root classes.

Finger Millet Root Hair QuantificationRoot hairs were observed to be widespread along the entirelengths of FM crown roots (Figures 5C,D; Supplementary FigureS1). There were no consistent significant differences in roothair length and density in FM in response to N limitation(Figures 5C,D; Supplementary Figure S1). However, in somecrown root tips, the newly initiating root hairs (on crown rootsegment 5) were either significantly shorter and/or less densewhen N-limited (Figures 5C,D; Supplementary Figure S1). Othertrends were observed: root hairs were generally longer and moredense in the upper (older) portions of the crown roots, whereasroot hairs became shorter and more sparsely distributed towardthe tips of the crown roots (Figures 5C,D). A similar trend wasobserved in 2012, though the differences were less pronounced(Supplementary Figure S1).

Physiological MeasurementsIn 2013, all plants were examined daily for evidence of flowering,defined as the point at which the emerging flower was firstobserved. Plants given +N began to flower at 85 DAT, while −Nplants did not begin flowering until 89 DAT. In 2012, the FMplants began flowering on the same day in both treatments.

In both 2012 and 2013, leaf chlorophyll was measured. In2012, the smaller FM plants showed no significant differencesin chlorophyll between treatments (Table 3). In 2013, with thelarger, more N-demanding plants, leaf chlorophyll (at 50 DAT) of−N plants was 56% lower than +N plants. At 80 DAT, the −N






TAB

LE

2|M

easu

rem

ents

of

roo

ttr

aits

of

FM

inre

spo

nse

tolo

wn

itro

gen

stre

ss.

Year

Dur

atio

nR

oo

tb

iom

ass

(g)∗

Cro

wn

roo

tnu

mb

erTo

talc

row

nro

ot

leng

th(m

m)

Ave

rag

ecr

ow

nro

ot

leng

th(m

m)

Tota

llat

eral

roo

tle

ngth

(mm

)To

talr

oo

tle

ngth

(mm

)La

tera

lro

ot:

cro

wn

roo

tra

tio

2012

+N14

2D

AT22

9.9

±13

.48,

n=

10A

80.8

±2.

65,n

=5

A64

90.2

±72

9.71

,n=

5A

79.7

±6.

87,n

=5

A20

926.

6±

1752

.15,

n=

5A

2741

5.8

±23

63.4

3,n

=5

A3.

3±

0.20

,n=

5A

−N14

2D

AT56

.2±

6.07

,n=

10B

69.6

±7.

00,n

=5

A63

65.4

±11

02.3

9,n

=5

A91

.9±

13.9

6,n

=5

A22

486.

1±

2676

.98,

n=

5A

2885

1.6

±37

48.9

6,n

=5

A3.

7±

0.25

,n=

5A

2013

+N13

9D

AT24

9.9

±37

.26,

n=

15a

216.

5±

27.2

5,n

=5

a11

134.

1±

1124

.29,

n=

4a

53.2

±1.

71,n

=5

a54

639.

1±

6949

.32,

n=

4a

6577

3.2

±79

57.0

9,n

=4

a4.

9±

0.30

,n=

4a

−N13

9D

AT13

6.2

±26

.25,

n=

13b

110.

8±

9.05

,n=

5b

6999

.8±

653.

04,n

=5

b64

.1±

6.53

,n=

5a

3457

6.1

±45

00.8

5,n

=5

b41

575.

9±

5123

.39,

n=

5b

4.9

±0.

26,n

=5

a

Lette

rsth

atar

edi

ffere

ntw

ithin

aye

arin

dica

teth

atth

em

eans

are

sign

ifica

ntly

diffe

rent

(P<

0.05

).D

ispl

ayed

valu

esar

em

ean±

SEM

.∗ R

oot

fresh

wei

ght

was

mea

sure

dbo

thye

ars.

FIGURE 3 | Shoot and root responses of FM to nitrogen limitation, atharvest. Representative mature, intact plants harvested in 2013, providedwith +N and −N treatments (A). The seed heads were pre-harvested. Thewhite scale bar at the lower right represents a length of 25 cm. Representativemature, entire root networks harvested in 2012 grown under +N and −Ntreatments, left and right respectively (B). The black scale bar at the lowerright represents a length of 10 cm.

treatment had a leaf chlorophyll content that was 23% lower than+N plants (Table 3). No significant differences were detectedin total N concentration of root, shoot, or seed head tissue(Table 3).

Discussion

Consistent with observations of subsistence farmers, our resultsshow that FM is a remarkable crop – surprisingly able toflower and produce healthy seed heads without any deliberatelyadded nitrogen (N). In our study, plants were grown in anartificial baked clay system with very poor N availability. N-poorenvironments are typical of the conditions under which this cropis grown in Sub-Saharan Africa and South Asia where no appliedN or residual N is used (National Research Council, 1996; Goronand Raizada, 2015). In South Asia, FM is often transplanted fromnurseries (National Research Council, 1996), a practice mimickedin this study.






FIGURE 4 | Correlations between FM shoot and seed head biomass androot architecture traits at harvest across nitrogen treatments. Number ofcrown roots vs. tillers (A). Shoot fresh weight vs. total root length (B), shootfresh weight vs. total crown root length (C), shoot fresh weight vs. number ofcrown roots (D), shoot fresh weight vs. calculated average crown root length(E), and shoot fresh weight vs. total lateral root length (F). Number of crown

roots vs. total crown root length (G). Seed head dry weight vs. total root length(H), seed head dry weight vs. total crown root length (I), seed head dry weightvs. number of crown roots (J), seed head dry weight vs. calculated averagecrown root length (K), and seed head dry weight vs. total lateral root length (L).Pearson r values are displayed. A single asterisk denotes significance atP < 0.10. A double asterisk denotes significance at P < 0.05.

The growth and acclimation of FM roots and root hairs inresponse to nutrient stress have not previously been reported.In this study, we have surveyed a range of responses employedby FM for growth in N-poor environments. Consistent withother cereal crop plants (Hay and Porter, 2006), FM shoot,

root, and seed head biomass decreased in the −N treatment,associated with declines in shoot tiller number, crown rootnumber, total crown root length and total lateral root length, butwith no consistent changes in root hair traits. These results aresummarized in a model (Figure 6).






FIGURE 5 | Root hair morphometric traits from 2013 plants at harvest.In 2013 and 2012 (data in Supplemental Figure S1), four different classes ofcrown roots were sampled based on their lengths/ages (A), and thenexamined for root hairs at five different segments spaced evenly along eachcrown root using light microscopy with 5x magnification (B). Root hairs werequantified for length (C) and density (D) at different segments of the crownroots (x-axis) ranging from the top/nearest the shoot (crown root segment 1)to near the root tip (crown root segment 5). A blue asterisk (∗) directly above

a mean data point denotes a significant difference in the root hair trait (atP < 0.05) between N treatments within an individual crown root segment.An asterisk above the green dashed line denotes a significant difference inthe root hair trait (at P < 0.05) between the top crown root segment(segment 1) and the bottom crown root segment (segment 2) within the +Ntreatment (red asterisk) or the −N treatment (black asterisk). All statisticalanalyses were performed with unpaired t-tests (with Welch’s correction whereunequal variances required).






Finger Millet may have Efficient NitrogenUptake AbilityThe baked-clay gravel (Turface R© ) in which plants were grownintrinsically contains very low N (Supplementary Table S2). Theclay gravel used contained 0.053% total N, which is substantiallylow compared to the surface layer of most cultivated soils inwhich N content can range between 0.06 and 0.5% (Bremner,1996; Johnston-Monje et al., 2014). Also, it is important to notethat the media utilized in these experiments was not soil, buthard, inert gravel, from which the majority of nitrogen would beunavailable for uptake during the life cycle of the FM plants. Totalnitrogen was measured by soaking the clay gravel in the N-freenutrient solution used in the experiments, and was found to be1.42 mg/L, or 0.1 mM total N. In rice and maize, these conditionswould be considered to cause N starvation (Schluter et al., 2012;Yang et al., 2015). To provide perspective, the United StatesEnvironmental Protection Agency defines clean drinking waterfor humans as containing no more than 10 mg/L of N as nitrate –a level more than seven times that which was available to FMplants over the course of our study2. Furthermore, our N valuesare likely overestimates, as they do not take into account leachingof N, given that the coarse clay had a low surface area and hencepoor N binding capacity, with large air pockets to facilitate rapiddownward nutrient flow away from the rhizosphere throughholes in the sides of the pails.

The question then arises as to how FM was able to reachmaturity with no deliberately addedN, and with such low levels ofN endogenous to the growth system. Furthermore, surprisingly,plants showed inconsistent differences in chlorophyll between the−N vs. +N treatments (Table 3). In 2013 the plants receivingthe −N treatment contained significantly less chlorophyll thanthe +N plants. This significant difference was not observedin 2012, although the trend remained similar. Additionally,

2http://water.epa.gov/drink/contaminants/basicinformation/nitrate.cfm#four

TABLE 3 | Finger millet tissue chlorophyll and nitrogen content inresponse to low nitrogen stress.

Year Measurement Treatment

2012 Chlorophyll content(SPAD)

+N 11.3 ± 1.27, n = 5 (106 DAT) A

−N 8.2 ± 1.43, n = 5 (106 DAT) A

2013 Chlorophyll content(SPAD)

+N 51.9 ± 4.06, n = 20 (80 DAT) a

−N 39.8 ± 3.07, n = 18 (80 DAT) b

2013 Shoot nitrogen content(% dry weight)

+N 1.9 ± 0.06, n = 3 (142 DAT) a

−N 2.1 ± 0.19, n = 3 (142 DAT) a

Root nitrogen content(% dry weight)

+N 1.3 ± 0.31, n = 3 (142 DAT) a

−N 0.9 ± 0.14, n = 3 (142 DAT) a

Seed head nitrogencontent (% dry weight)

+N 2.5 ± 0.17, n = 3 (142 DAT) a

−N 2.1 ± 0.17, n = 3 (142 DAT) a

Letters that are different within a year indicate that the means are significantlydifferent (P < 0.05).

FIGURE 6 | Summary of the effects of low N stress on FM architecture.Low N resulted in a reduction in end-season shoot biomass and seed headbiomass associated with fewer tillers compared to plants receiving higher N(A). Fewer crown roots were initiated in 2013 in response to the −N treatment(B), likely responsible for the reduction in total lateral root system length (C).The calculated average crown root length was unchanged between nitrogentreatments. No consistent changes in root hair length or density occurred forroot hairs along the crown roots in response to the −N treatment (D).However, in both N treatments, root hairs were generally longer and denser atthe top of the crown roots than at the bottom of the crown roots (D).

there were no significant differences in tissue N concentration(Table 3). In other cereal crops, N starvation has been shownto cause consistent declines in chlorophyll (Széles et al., 2012;Muñoz-Huerta et al., 2013; Wang et al., 2014) and substantialdeclines in tissue N concentration (Havlin et al., 1999; Chenet al., 2015). Here, the plants were grown on an outdoor field inopen pails (Figures 1B,C), closed at the bottom but perforatedwith four drainage holes on the side, very close to the bottomedge (Supplementary Figure S2). We did sometimes observeroots growing from these drainage holes, but only in olderplants. It is possible that N was taken up by these late stageroots to assist with grain fill, but nevertheless, the plants wouldhave needed to grow for months without these hypotheticalscavenging roots, using starter N from an extremely small seed.Furthermore, we did not observe weeds in the pails (e.g., that


http://water.epa.gov/drink/contaminants/basicinformation/nitrate.cfm#four





might host N-fixing microbes) nor an obvious mycorrhizalnetwork at root harvesting and during root imaging. Some Nmay have entered from unintended sources such as from animalactivity (we observed bird feathers), N-fixation by lightning,algae (growth was observed in pots), or microbial growth in thefertigation lines. These external N inputs were likely minisculeand suggest that FM roots themselves may have excellent Nuptake ability.

In the future, it may be worth investigating whether FM hostsN-fixing microbes, as is observed in its grass relative, sugarcane(Cavalcante and Dobereiner, 1988; James et al., 1994). Weconducted a preliminary experiment to test whether culturablemicrobes from FM root and shoot extracts can grow on agarwithout N, but no colonies appeared (data not shown).

These results also provide further impetus to examine FMas a source of novel agronomic traits for genetic dissection atthe molecular level, especially with regards to N uptake andutilization efficiency mechanisms. For example, the behavior ofDof1 and Dof2, transcription factors implicated in controllingmany responses to N, have already been characterized in FMunder conditions of varying N applications (Gupta et al., 2014,2015; Kumar et al., 2014). However, other molecular geneticstudies concerning N in FM are scarce.

In the –N Treatment, Shoot, and Total SeedHead Biomass Decreases were Associatedwith Declines in Tiller NumberConsistent with other cereal crop plants (Hay and Porter, 2006),FM shoot and total seed head biomass decreased in the –Ntreatment (Table 1), which were associated with declines in tillernumber in both 2012 and 2013. Inadequate N supply has longbeen known to lower biomass and yield in plants that display tillerplasticity including wheat, barley, rice, several bioenergy grasses,and teosinte, the progenitor tomodernmaize (Mae, 1997; Gaudinet al., 2011b; Alzueta et al., 2012; Waramit et al., 2014).

In the –N Treatment in 2013, Root BiomassDeclines were Associated with DiminishedRoot TraitsIn other cereals including maize, low N availability was shownto increase the length of the root system, decrease the crownroot number, and increase the lateral root to crown root ratio(Eghball and Maranville, 1993; Wang et al., 2004; Chun et al.,2005; Liu et al., 2009; Gaudin et al., 2011a). However, in FM plantsharvested in 2013, we observed a decrease in the total lengthof both the lateral root system and crown root system in theabsence of N fertilizer, as well as a decrease in crown root number(Table 2). Lateral root to crown root ratios remained statisticallyequivalent between treatments. In 2012 root architecture did notdiffer significantly between treatments.

Differences between our observations and the literature mightbe explained by the fact that the earlier studies using othercereals were conducted under conditions of N limitation, whileour FM plants may have experienced extreme N stress (i.e.,0.1 mM N from the clay growth medium). The majority ofexperiments thus far investigating plant root responses to trueN starvation have been conducted in Arabidopsis. A group

of Arabidopsis ecotypes has been identified which respond toN starvation with decreases in various root traits includingroot fresh biomass (Ikram et al., 2012). However, it shouldbe noted that the majority of these ecotypes responded byincreasing root biomass, indicating there is a substantial degreeof genetic diversity in plant root acclimation responses. Giventhese prior results, it will be helpful to phenotype the rootsof multiple genotypes of FM in response to extreme Nstress.

Research into the responses of crop plants to N starvation,including inmaize and barley, is primarily restricted to molecularanalysis, with a focus on the regulation of N assimilationgenes (Lee et al., 1992; Møller et al., 2011; Nacry et al., 2013).Additionally, because of the difficulties of imposing true Nstarvation on nutrient-intensive crops, the stressful condition isoften applied only for a short time, unlike the current study inwhich it was the entire duration of the FM life cycle.

Finger Millet Displays Whole-PlantCoordination of Architectural Traits inResponse to Low NTiller number and crown root number were found to becorrelated across the N treatments (Figure 4), indicating thatFM’s above ground and below ground morphologies may betightly linked. To ensure that this relationship is robust, similarcorrelations should be generated in the future with additionalN treatments (see below). In modern maize, the transcriptionfactor Teosinte Branched 1 (TB1) is in part responsible for therepression of tiller outgrowth, and results in the plant’s single-stem growth habit (Doebley et al., 1995; Lukens and Doebley,2001). It has been shown that selection pressure during thedomestication of maize from teosinte led to changes in expressionof the Tb1 gene and coordinated declines in tiller and crown rootnumber (Hubbard et al., 2002; Gaudin et al., 2011b, 2014). AsFM is genetically related to maize, a similar mechanism may bedirecting the tight correlation between the number of tillers andcrown roots and it may be of interest to investigate the presenceof Tb1 orthologs in the FM genome.

Crown Root Initiation may be the PrimaryDeterminant of Changes in Root Architectureto Support Altered Shoot GrowthAlthough the relationship between FM yield and above groundtraits including plant height and shoot biomass has beenpreviously investigated (Wolie andDessalegn, 2011), this study is,to the best of our knowledge, the first time that such correlationshave been performed on below ground plant architecture in FM.The highest correlations observed between FM shoot and seedhead biomass and root traits were with the crown root number,not the average crown root length (Figure 4). Furthermore,of the various root traits that changed in response to the-N treatment, the crown root number showed the greatestdecline (−49%). Total crown root length was found to correlatetightly with crown root number (Figure 4G). Since lateralroots originate from crown roots, then combined these resultsindicate that declines in crown root initiation in response toN limitation primarily determine the other root correlations






observed (Figure 4). It seems logical that when plants haveincreased access to N, they respond by initiating additionalcrown root systems (with their lateral roots and root hairs)in order to support increased shoot biomass and seed headproduction.

The correlations generated here may aid future breedingefforts and trait association studies, in which below groundcontributors to yield and biomass could be explored. However,a relatively small amount of data was used to generate thesecorrelations, and a degree of prudence must be used in theinterpretation of the results. In particular, the consistency ofresponses within our +N and −N treatments resulted inclustered data (Figure 4). In future experiments, it may be usefulto add additional N treatments to make the correlation analysesmore robust.

As more of the FM genome sequence becomes available itmay be helpful to investigate orthologs of genes implicated incrown root initiation in other cereals including Tb1 (noted above)and RTCS identified in maize (Hubbard et al., 2002; Taraminoet al., 2007), as well as Crown rootless 1, 4, 5 and WOX11identified in rice (Inukai et al., 2005; Zhao et al., 2009; Coudertet al., 2010; Kitomi et al., 2011). Additional candidate geneticelements include those responsive to auxin or cytokinin signaling,pathways recognized as contributing to the control of crown rootgrowth and development (Inukai et al., 2005; Coudert et al., 2010;Kitomi et al., 2011).

Low N did not Induce Consistent Changes inRoot Hair Length or DensityBased on hand tracing of 14,420 root hairs in this study, weobserved no consistent differences in root hair length and densityin FM in response to the−N treatment (Figure 5; SupplementaryFigure S1). In the literature, the analysis of root hairs in responseto N limitation/starvation in crop plants is understudied, withcontradictory results. Root hairs in Arabidopsis and severalgrass species (Holcus lanatus L., Deschampsia flexuosa L., Poaannua L., and Lolium perenne L.) showed increased length anddensity in response to N limitation (Robinson and Rorison,1987; Bloch et al., 2011). Root hairs were hypothesized tocontain a mechanism for sensing low N stress (Shin et al.,2005). In maize and teosinte, however, root hairs showed declinesin length and density as a result of low N conditions usingan aeroponics misting system (Gaudin et al., 2011a,b). Onepotentially interesting observation from the present study isthat FM root hairs may have been longer and more densein response to low N at the tips of some root age classes(Figures 5C,D; Supplementary Figure S1A). In the future, preciseexamination of root hair initiation may help to confirm theseobservations.

Study LimitationsTo the best of our knowledge, this study is the first attemptto grow FM in the northern latitudes of Canada. A growingseason which is cooler and shorter than that of FM’s nativeEast Africa and India (Goron and Raizada, 2015) created a setof challenges and limitations. For example, it was necessary todelay germination and transplantation in the spring until local

temperatures could sustain growth. This resulted in importantdifferences between the 2012 and 2013 growing seasons.Although the total time between transplantation and harvest wassimilar each year (142 DAT in 2012 vs. 139 DAT in 2013), in 2012,transplantation was performed 18 days later and growth extendedinto November which borders the cold, winter season. In 2012,the late-season conditions became especially sub-optimal for FMwhich requires average daily minimum temperatures to be above18◦C for optimum growth3; the average temperature on the day ofharvest was −3.3◦C4. Plants were stunted with a greatly reducedtiller number in 2012 compared to the 2013 plants (Table 1), andas a result the plants were from a different growth stage at the timeof harvest (Figures 2B,D). Indeed, many biomass measurements(Tables 1 and 2) were significantly lower in 2012 than 2013,although the trends remained similar in terms of the responseto N treatments. Lower biomass in 2012 may also have beenthe result of pathogenesis as disease-like spots were observedat the seedling stage (they completely disappeared within a fewweeks); these symptoms were absent in 2013. As FM is anindeterminate plant, the early onset of cold in 2012 (comparedto 2013) combined with the slow growth of seedlings (perhapsdue to the disease) might have prevented the initiation of newtillers and roots in 2012. The cold likely also affected seed heads(Table 1), as both grain fill and vegetative growth occur in parallelin FM. Finally, these differences in growing conditions, combinedwith differences when measurements were taken relative to thegrowth stage between years, may have also contributed to theinconsistent chlorophyll readings discussed above.

In 2012, the decrease in root biomass from 230 to 56 g inthe −N treatment was not matched by large differences in thearchitectural traits (Table 2). In 2012, we were especially cautiouswhile washing the root systems after harvest to preserve root hairsand lateral roots. It is likely that fine Turface R© clay stuck to theroots as a result, especially on the larger +N root systems whichwere more dense and difficult to wash, resulting in a non-linearincrease in the biomass. In 2013, the roots were washed morethoroughly.

Another study constraint was that, although the pails used togrow plants were very large, they may have limited the growthof the FM root systems, especially in the +N treatment, thusreducing the significance of the differences betweenN treatments.A future experiment involving growing plants directly in fieldsoil could be informative, but it would not be possible to imposeextreme N stress conditions and would be especially challengingto phenotype fine root traits.

A final study limitation was that in 2012, shoot dry weightwas measured while in 2013 shoot fresh weight was measured,creating a challenge to directly compare biomass results betweenyears (Table 1). As the leaf tissue in grass species is mostlywater (Garnier, 1992; Garnier and Laurent, 1994), the differencesbetween the shoot biomass data in 2012 vs. 2013 are notunreasonable. In any case, all comparisons were made withinyears to circumvent this issue. For seed heads and roots, however,consistent measurements were used across both years.

3http://www.cabi.org/cpc/datasheet/20674#tab1-nav4http://climate.weather.gc.ca/


http://www.cabi.org/cpc/datasheet/20674#tab1-nav

http://climate.weather.gc.ca/





Study Implications and Future ExperimentsThis study has provided the first detailed description of FMresponses to extreme N stress, including the first description ofroot acclimation responses. The results suggest that FM rootsmay have extreme N uptake ability. Mapping of the underlyinggenes may aid marker assisted selection (MAS) based efforts toimprove NUE in other cereal crops.

Only a single genotype was examined in this study; however,there are many landraces available for FM from Africa and SouthAsia that show diverse adaptations to local climates (Jarvis et al.,2008; Wolie and Dessalegn, 2011; Goron and Raizada, 2015). Itwill be useful to compare the results from this study to a diversitypanel of FM to identify those with particularly high NUE anduptake ability. Through such guided breeding approaches, it maybe possible to help subsistence farmers in Africa and South Asiawho have poor access to N fertilizer.

Author Contributions

TLG undertook all lab experiments, the 2013 field experiments,performed all analyses and wrote the manuscript. VKB helped toconceive the study, and designed and set up the field experimentin 2012. CRS helped to set up the field experiments and assistedwith plant growth and measurements. SW assisted with root hairmicroscopy and morphometric measurements. MNR conceivedof the study and edited the manuscript.

Acknowledgments

We thank Kirit Patel (Canadian Mennonite University,Canada) for inspiring our interest in FM. We thank EmmaHarris (University of Guelph) for painstaking assistance inscanning roots. We thank Amelie Gaudin (University ofCalifornia, Davis) for valuable advice in root hair scanningand microscopy. Hugo Gonzalez (University of Guelph) gaveimportant direction in the operation of the Turface R© fertigationgrowth system. We thank Clarence Swanton (Universityof Guelph) for use of the fertigation system used in thisexperiment. We thank Tadros Atalla, Marina Atalla, and AlexWhittal for assistance with field harvesting. We thank DylanHarding (University of Guelph) for reading and editing themanuscript. TG was supported in part by scholarships fromthe University of Guelph, and a QEII-GSST award fromthe government of Ontario. This research was supported byCIFSRF grants to MNR from the Canadian InternationalDevelopment Research Centre (IDRC) and the CanadianDepartment of Foreign Affairs, Trade and Development(DFATD).

Supplementary Material

The Supplementary Material for this article can be found onlineat: http://journal.frontiersin.org/article/10.3389/fpls.2015.00652

References

Alzueta, I., Abeledo, L. G., Mignone, C. M., and Miralles, D. J. (2012).Differences between wheat and barley in leaf and tillering coordination undercontrasting nitrogen and sulfur conditions. Eur. J. Agron. 41, 92–102. doi:10.1016/j.eja.2012.04.002

Bhoite, S. V., and Nimbalkar, V. S. (1996). Response of finger millet cultivars tonitrogen and phosphorus under rain-fed conditions. J. MaharashtraAgric. Univ.20, 189–190.

Bloch, D., Monshausen, G., Singer, M., Gilroy, S., and Yalovsky, S.(2011). Nitrogen source interacts with ROP signalling in root hair tip-growth. Plant Cell Environ. 34, 76–88. doi: 10.1111/j.1365-3040.2010.02227.x

Bremner, J. M. (1996). “Nitrogen-total,” in Methods of Soil Analysis. Part 3.Chemical Methods, ed. J. M. Bartels (Madison, WI: Soil Society of America andAmerican Society of Agronomy), 1085–1091.

Cavalcante, V. A., and Dobereiner, J. (1988). A new acid-tolerant nitrogen-fixing bacterium associated with sugarcane. Plant Soil 108, 23–31. doi:10.1007/BF02370096

Chapin, F. S., Bloom, A. J., Field, C. B., and Waring, R. H. (1987). Plant responsesto multiple environmental factors. Bioscience 37, 49–57. doi: 10.2307/1310177

Chen, Y., Xiao, C., Wu, D., Xia, T., Chen, Q., Chen, F., et al. (2015). Effects ofnitrogen application rate on grain yield and grain nitrogen concentration intwo maize hybrids with contrasting nitrogen remobilization efficiency. Eur. J.Agron. 62, 79–89. doi: 10.1016/j.eja.2014.09.008

Chun, L., Mi, G., Li, J., Chen, F., and Zhang, F. (2005). Genetic analysis of maizeroot characteristics in response to low nitrogen stress. Plant Soil 276, 369–382.doi: 10.1007/s11104-005-5876-2

Ciampitti, I. A., Murrell, S. T., Camberato, J. J., Tuinstra, M., Xia, Y.,Friedemann, P., et al. (2013). Physiological dynamics of maize nitrogen uptakeand partitioning in response to plant density and N stress factors: I. vegetativephase. Crop Sci. 53, 2105–2119. doi: 10.2135/cropsci2013.01.0040

Coudert, Y., Périn, C., Courtois, B., Khong, N. G., and Gantet, P. (2010). Geneticcontrol of root development in rice, the model cereal. Trends Plant Sci. 15,219–226. doi: 10.1016/j.tplants.2010.01.008

DeBruin, J., and Butzen, S. (2014). Nitrogen uptake in corn. Pioneer CropInsights 24.

Dida, M. M., Wanyera, N., Harrison, M. L., Bennetzen, J. L., and Devos,K. M. (2008). Population structure and diversity in finger millet (Eleusinecoracana) germplasm. Trop. Plant Biol. 1, 131–141. doi: 10.1007/s12042-008-9012-3

Doebley, J., Stec, A., and Gustus, C. (1995). Teosinte branched1 and the originof maize: evidence for epistasis and the evolution of dominance. Genetics 141,333–346.

Dorosh, P. A., Dradri, S., and Haggblade, S. (2009). Regional trade, governmentpolicy and food security: recent evidence from Zambia. Food Policy 34, 350–366.doi: 10.1016/j.foodpol.2009.02.001

Dubley, O. P., and Shrivas, B. N. (1999). Response of finger millet (Eleusinecoracana) genotypes to nitrogen. Indian J. Agron. 44, 564–566.

Eghball, B., and Maranville, J. W. (1993). Root development and nitrogen influx ofcorn genotypes grown under combined drought and nitrogen stresses.Agron. J.85, 147–152. doi: 10.2134/agronj1993.00021962008500010027x

Etheridge, R. D., Pesti, G. M., and Foster, E. H. (1998). A comparison ofnitrogen values obtained utilizing the Kjeldahl nitrogen and Dumas combustionmethodologies (Leco CNS 2000) on samples typical of an animal nutritionanalytical laboratory. Anim. Feed Sci. Technol. 73, 21–28. doi: 10.1016/S0377-8401(98)00136-9

Fiedler, R., Proksch, G., and Koepf, A. (1973). The determination of total nitrogenin plant materials with an automatic nitrogen analyser. Anal. Chim. Acta 63,435–443. doi: 10.1016/S0003-2670(01)82368-0

Garnier, E. (1992). Growth analysis of congeneric annual and perennial grassspecies. J. Ecol. 80, 665–675. doi: 10.2307/2260858

Garnier, E., and Laurent, G. (1994). Leaf anatomy, specific mass and water contentin congeneric annual and perennial grass species. New Phytol. 128, 725–736.doi: 10.1111/j.1469-8137.1994.tb04036.x


http://journal.frontiersin.org/article/10.3389/fpls.2015.00652





Gaudin, A. C. M., McClymont, S. A., Holmes, B. M., Lyons, E., and Raizada,M. N. (2011a). Novel temporal, fine-scale and growth variation phenotypesin roots of adult-stage maize (Zea mays L.) in response to low nitrogenstress. Plant Cell Environ. 34, 2122–2137. doi: 10.1111/j.1365-3040.2011.02409.x

Gaudin, A. C. M., McClymont, S. A., and Raizada, M. N. (2011b). The nitrogenadaptation strategy of the wild teosinte ancestor of modern maize, Zeamays subsp. parviglumis. Crop Sci. 51, 2780–2795. doi: 10.2135/cropsci2010.12.0686

Gaudin, A. C. M., McClymont, S. A., Soliman, S. S. M., and Raizada, M. N.(2014). The effect of altered dosage of a mutant allele of Teosinte branched1 (tb1-ref) on the root system of modern maize. BMC Genet. 15:23. doi:10.1186/1471-2156-15-23

Goron, T. L., and Raizada, M. N. (2015). Genetic diversity and genomic resourcesavailable for the small millet crops to accelerate a NewGreen Revolution. Front.Plant Sci. 6:157. doi: 10.3389/fpls.2015.00157

Goron, T. L., Watts, S., Shearer, C., and Raizada, M. N. (2015). Growth in Turfaceclay permits root hair phenotyping along the entire crown root in cereal cropsand demonstrates that root hair growth can extend well beyond the root hairzone. BMC Res. Notes 8:143. doi: 10.1186/s13104-015-1108-x

Gruère, G., Nagarajan, L., and King, E. D. I. O. (2009). The role of collectiveaction in the marketing of underutilized plant species: lessons from acase study on minor millets in South India. Food Policy 34, 39–45. doi:10.1016/j.foodpol.2008.10.006

Gupta, A. K., Guar, V. S., Gupta, S., and Kumar, A. (2013).Nitrate signals determinethe sensing of nitrogen through differential expression of genes involved innitrogen uptake and assimilation in finger millet. Funct. Integr. Genomics 13,179–190. doi: 10.1007/s10142-013-0311-x

Gupta, N., Gupta, A. K., Gaur, V. S., and Kumar, A. (2012). Relationship of nitrogenuse efficiency with the activities of enzymes involved in nitrogen uptake andassimilation of finger millet genotypes grown under different nitrogen inputs.Sci. World J. 2012, 1–10. doi: 10.1100/2012/625731

Gupta, S., Gupta, S.M., Gupta, A. K., Gaur, V. S., and Kumar, A. (2014). Fluctuationof Dof1/Dof2 expression ratio under the influence of varying nitrogen and lightconditions: involvement in differential regulation of nitrogen metabolism intwo genotypes of finger millet (Eleusine coracana L.). Gene 546, 327–335. doi:10.1016/j.gene.2014.05.057

Gupta, S., Malviya, N., Kushwaha, H., Nasim, J., Bisht, N. C., Singh, V. K., et al.(2015). Insights into structural and functional diversity of Dof (DNA bindingwith one finger) transcription factor. Planta 241, 549–562. doi: 10.1007/s00425-014-2239-3

Havlin, J., Beaton, J., Tisdale, S., and Nelson, W. (1999). “Nitrogen,” in Soil Fertilityand Fertilizers, 6th Edn, eds M. Carnis and L. Harvey (Upper Saddle River, NJ:Prentice-Hall, Inc).

Hay, R., and Porter, J. (2006). “Limiting factors and the achievement of highyield,” in The Physiology of Crop Yield (Oxford: Blackwell Publishing Inc.),180–204.

Hubbard, L., McSteen, P., Doebley, J., and Hake, S. (2002). Expression patterns andmutant phenotype of teosinte branched1 correlate with growth suppression inmaize and teosinte.Genetics 162, 1927–1935.

Ikram, S., Bedu, M., Daniel-Vedele, F., Chaillou, S., and Chardon, F. (2012).Natural variation of Arabidopsis response to nitrogen availability. J. Exp. Bot.63, 91–105. doi: 10.1093/jxb/err244

Inukai, Y., Sakamoto, T., Ueguchi-Tanaka, M., Shibata, Y., Gomi, K., Umemura, I.,et al. (2005). Crown rootless1, which is essential for crown root formation inrice, is a target of an AUXIN RESPONSE FACTOR in auxin signaling. PlantCell 17, 1387–1396. doi: 10.1105/tpc.105.030981

James, E., Reis, V., Olivares, F., Baldani, J., and Döbereiner, J. (1994). Infection ofsugar cane by the nitrogen-fixing bacterium Acetobacter diazotrophicus. J. Exp.Bot. 45, 757–766. doi: 10.1093/jxb/45.6.757

Jarvis, D. I., Brown, A. H. D., Cuong, P. H., Collado-panduro, L., Latournerie-moreno, L., Gyawali, S., et al. (2008). A global perspective of therichness and evenness of tradtitional crop-variety diversity maintained byfarming communities. Proc. Natl. Acad. Sci. U.S.A. 105, 5326–5331. doi:10.1073/pnas.0800607105

Johnston-Monje, D., Mousa, W., Lazarovits, G., and Raizada, M. N. (2014). Impactof swapping soils on the endophytic bacterial communities of pre-domesticated,ancient and modern maize. BMC Plant Biol. 14:233. doi: 10.1186/s12870-014-0233-3

Keeney, D. R., and Bremner, J. M. (1966). Comparison and evaluation oflabroratory methods of obtaining an index of soil nitrogen availability. Agron. J.58, 498–503. doi: 10.2134/agronj1966.00021962005800050013x

Kitomi, Y., Ito, H., Hobo, T., Aya, K., Kitano, H., and Inukai, Y. (2011). Theauxin responsive AP2 / ERF transcription factor CROWN ROOTLESS5 isinvolved in crown root initiation in rice through the induction of OsRR1,a type-A response regulator of cytokinin signaling. Plant J. 3, 472–484. doi:10.1111/j.1365-313X.2011.04610.x

Kumar, A., Kanwal, P., Gupta, A. K., Singh, B. R., and Gaur, V. S. (2014). A full-length Dof1 transcription factor of finger millet and its response to a circadiancycle. Plant Mol. Biol. Report. 32, 419–427. doi: 10.1007/s11105-013-0653-5

Lee, R. B., Purves, J. V., Ratcliffe, R. G., and Saker, L. R. (1992). Nitrogenassimilation and the control of ammonium and nitrate absorption by maizeroots. J. Exp. Bot. 43, 1385–1396. doi: 10.1093/jxb/43.11.1385

Liu, J., Chen, F., Olokhnuud, C., Glass, A. D. M., Tong, Y., Zhang, F., et al. (2009).Root size and nitrogen-uptake activity in two maize (Zea mays) inbred linesdiffering in nitrogen-use efficiency. J. Plant Nutr. Soil Sci. 172, 230–236. doi:10.1002/jpln.200800028

Lukens, L., and Doebley, J. (2001). Molecular evolution of the teosinte branchedgene among maize and related grasses. Mol. Biol. Evol. 18, 627–638. doi:10.1093/oxfordjournals.molbev.a003843

Mae, T. (1997). Physiological nitrogen efficiency in rice: nitrogen utilization,photosynthesis, and yield potential. Plant Soil 196, 201–210. doi:10.1023/A:1004293706242

Møller, A. L. B., Pedas, P., Andersen, B., Svensson, B., Schjoerring, J. K., andFinnie, C. (2011). Responses of barley root and shoot proteomes to long-termnitrogen deficiency, short-term nitrogen starvation and ammonium. Plant CellEnviron. 34, 2024–2037. doi: 10.1111/j.1365-3040.2011.02396.x

Moll, R. H., Kamprath, E. J., and Jackson, W. A. (1982). Analysis and interpretationof factors which contribute to efficiency of nitrogen-utilization. Agron. J. 74,562–564. doi: 10.2134/agronj1982.00021962007400030037x

Muñoz-Huerta, R. F., Guevara-Gonzalez, R. G., Contreras-Medina, L. M., Torres-Pacheco, I., Prado-Olivarez, J., and Ocampo-Velazquez, R. V. (2013). A reviewof methods for sensing the nitrogen status in plants: advantages, disadvantagesand recent advances. Sensors 13, 10823–10843. doi: 10.3390/s130810823

Nacry, P., Bouguyon, E., and Gojon, A. (2013). Nitrogen acquisition by roots:physiological and developmental mechanisms ensuring plant adaptation to afluctuating resource. Plant Soil 370, 1–29. doi: 10.1007/s11104-013-1645-9

National Research Council. (1996). Lost Crops of Africa.Washington, DC: NationalAcademy Press.

Page, E. R., Liu, W., Cerrudo, D., Lee, E. A., and Swanton, C. J. (2011). Shadeavoidance influences stress tolerance in maize. Weed Sci. 59, 326–334. doi:10.1614/WS-D-10-00159.1

Pradhan, A., Thakur, A., Patel, S., andMishra, N. (2011). Effect of different nitrogenlevels on kodo and finger millet under rainfed conditions. Res. J. Agric. Sci. 2,136–138.

Puttaswamy, S., and Krishnamurthy, K. (1976). Relative dry-matter efficiency infinger millet genotypes in relation to levels of spacing and nitrogen. Mysore J.Agr. Sci. 10, 345–352.

Robinson, D., and Rorison, I. H. (1987). Root hairs and plant growth atlow nitrogen availabilities. New Phytol. 107, 681–693. doi: 10.1111/j.1469-8137.1987.tb00906.x

Roy, D., Chakraborty, T., Sounda, G., and Maitra, S. (2001). Effect of fertilitylevels and plant population on yield and uptake of nitrogen, phosphorus andpotassium in finger millet (Eleusine coracana) in lateritic soil of West Bengal.Indian J. Agron. 464, 707–711.

Schluter, U., Mascher, M., Colmsee, C., Scholz, U., Brautigam, A., Fahnenstich, H.,et al. (2012). Maize source leaf adaptation to nitrogen deficiency affects not onlynitrogen and carbon metabolism but also control of phosphate homeostasis.Plant Physiol. 160, 1384–1406. doi: 10.1104/pp.112.204420

Shin, R., Berg, R. H., and Schachtman, D. P. (2005). Reactive oxygen species androot hairs in Arabidopsis root response to nitrogen, phosphorus and potassiumdeficiency. Plant Cell Physiol. 46, 1350–1357. doi: 10.1093/pcp/pci145

Széles, A. V., Megyes, A., and Nagy, J. (2012). Irrigation and nitrogen effects on theleaf chlorophyll content and grain yield of maize in different crop years. Agric.Water Manag. 107, 133–144. doi: 10.1016/j.agwat.2012.02.001

Taramino, G., Sauer, M., Stauffer, J. L. S. Jr., Multani, D., Niu, X., and Sakai, H.(2007). The maize (Zea mays L.) RTCS gene encodes a LOB domain protein






that is a key regulator of embryonic seminal and post-embryonic shoot-borneroot initiation. Plant J. 1, 649–659. doi: 10.1111/j.1365-313X.2007.03075.x

Thilakarathna, M. S., and Raizada, M. N. (2015). A review of nutrient managementstudies involving finger millet in the semi-arid tropics of Asia and Africa.Agronomy 5, 262–290. doi: 10.3390/agronomy5030262

Tollenaar, M., and Migus, W. (1984). Dry matter accumulation of maize grownhydroponically under controlled-environment and field conditions. Can. J.Plant Sci. 64, 475–485. doi: 10.4141/cjps84-070

Upadhyaya, H. D., Gowda, C. L. L., Pundir, R. P. S., Reddy, V. G., andSingh, S. (2006). Development of core subset of finger millet germplasm usinggeographical origin and data on 14 quantitative traits. Genet. Resour. Crop Evol.53, 679–685. doi: 10.1007/s10722-004-3228-3

Wang, G., Bronson, K. F., Thorp, K. R., Mon, J., and Badaruddin, M.(2014). Multiple leaf measurements improve effectiveness of chlorophyllmeter for durum wheat nitrogen management. Crop Sci. 54, 817–826. doi:10.2135/cropsci2013.03.0160

Wang, Y., Mi, G., Chen, F., and Zhang, J. (2004). Response of root morphology tonitrate supply and its contribution to nitrogen accumulation in maize. J. PlantNutr. 27, 2189–2202. doi: 10.1081/LPLA-200034683

Waramit, N., Moore, K. J., and Heaton, E. (2014). Nitrogen and harvest date affectdevelopmental morphology and biomass yield of warm-season grasses. GCBBioenergy 6, 534–543. doi: 10.1111/gcbb.12086

Wolie, A., and Dessalegn, T. (2011). Correlation and path coefficient analysesof some yield related traits in finger millet ( Eleusine coracana (L.) Gaertn.)germplasms in northwest Ethiopia. Afr. J. Agric. Res. 6, 5099–5105. doi:10.5897/AJAR11.616

Yang, W., Yoon, J., Choi, H., Fan, Y., Chen, R., and An, G. (2015). Transcriptomeanalysis of nitrogen-starvation- responsive genes in rice. BMC Plant Biol. 15:31.doi: 10.1186/s12870-015-0425-5

Zhao, Y., Hu, Y., Dai, M., Huang, L., and Zhou, D.-X. (2009). The WUSCHEL-related homeobox gene WOX11 is required to activate shoot-borne crown rootdevelopment in rice. Plant Cell 21, 736–748. doi: 10.1105/tpc.108.061655

Conflict of Interest Statement: The authors declare that the research wasconducted in the absence of any commercial or financial relationships that couldbe construed as a potential conflict of interest.

Copyright © 2015 Goron, Bhosekar, Shearer, Watts and Raizada. This is an open-access article distributed under the terms of the Creative Commons AttributionLicense (CC BY). The use, distribution or reproduction in other forums is permitted,provided the original author(s) or licensor are credited and that the originalpublication in this journal is cited, in accordance with accepted academic practice.No use, distribution or reproduction is permitted which does not comply with theseterms.










whole plant acclimation responses by finger millet to low ...€¦ · manish n. raizada* department...

Documents